CN103338403B - Individual character program commending method in radio data system and this system - Google Patents

Individual character program commending method in radio data system and this system Download PDF

Info

Publication number
CN103338403B
CN103338403B CN201310272576.9A CN201310272576A CN103338403B CN 103338403 B CN103338403 B CN 103338403B CN 201310272576 A CN201310272576 A CN 201310272576A CN 103338403 B CN103338403 B CN 103338403B
Authority
CN
China
Prior art keywords
mentioned
rating
bunch
crowd
program
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201310272576.9A
Other languages
Chinese (zh)
Other versions
CN103338403A (en
Inventor
殷复莲
柴剑平
高雅
张韬政
覃垚
江茜
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Communication University of China
Original Assignee
Communication University of China
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Communication University of China filed Critical Communication University of China
Priority to CN201310272576.9A priority Critical patent/CN103338403B/en
Publication of CN103338403A publication Critical patent/CN103338403A/en
Application granted granted Critical
Publication of CN103338403B publication Critical patent/CN103338403B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
    • H04N21/466—Learning process for intelligent management, e.g. learning user preferences for recommending movies
    • H04N21/4662—Learning process for intelligent management, e.g. learning user preferences for recommending movies characterized by learning algorithms
    • H04N21/4665—Learning process for intelligent management, e.g. learning user preferences for recommending movies characterized by learning algorithms involving classification methods, e.g. Decision trees
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
    • H04N21/458—Scheduling content for creating a personalised stream, e.g. by combining a locally stored advertisement with an incoming stream; Updating operations, e.g. for OS modules ; time-related management operations
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
    • H04N21/25—Management operations performed by the server for facilitating the content distribution or administrating data related to end-users or client devices, e.g. end-user or client device authentication, learning user preferences for recommending movies
    • H04N21/258—Client or end-user data management, e.g. managing client capabilities, user preferences or demographics, processing of multiple end-users preferences to derive collaborative data
    • H04N21/25866—Management of end-user data
    • H04N21/25891—Management of end-user data being end-user preferences
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
    • H04N21/466—Learning process for intelligent management, e.g. learning user preferences for recommending movies
    • H04N21/4667—Processing of monitored end-user data, e.g. trend analysis based on the log file of viewer selections
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
    • H04N21/466—Learning process for intelligent management, e.g. learning user preferences for recommending movies
    • H04N21/4668—Learning process for intelligent management, e.g. learning user preferences for recommending movies for recommending content, e.g. movies
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/47—End-user applications
    • H04N21/472—End-user interface for requesting content, additional data or services; End-user interface for interacting with content, e.g. for content reservation or setting reminders, for requesting event notification, for manipulating displayed content

Abstract

The invention provides a kind of radio data system, comprising: input part, carrying out various parameter needed for individual character program commending and various instruction for inputting above-mentioned radio data system; Programme information storage part, for storing information about various broadcast TV program and data; Analytic unit, is utilized the various parameter inputted by input part and the information about broadcast TV program read from above-mentioned programme information storage part, generates the individual character programme that will send, and determine the recommendation crowd as sending object; And recommendation information sending part, to the above-mentioned individual character programme of above-mentioned rating people pocket transmission determined.According to this structure, to the different user of radio data system, different individual character programmes can be recommended.

Description

Individual character program commending method in radio data system and this system
Technical field
The present invention relates in broadcasting television technology field, in more detail, relate to the individual character program commending method can recommended respectively to specific rating colony in the radio data system of corresponding individual character program and this radio data system.
Background technology
Along with TV programme becomes increasingly abundant, radio and television user is faced with the problem of with Internet user similar " information overload ", under such circumstances, how can follow the tracks of the interests change of user, the problem finding the interested content of TV program of user is very urgent, and radio and television individual character program recommendation system can effectively address this problem.
The theoretical foundation of radio and television individual character program commending is decision support technique and data mining technology.DSS (DSS, DecisionSupportSystem) is proposed in 20 century 70s first by American scientist Keen and ScottMorton, has achieved huge development to the eighties in 20th century.Along with the continuous research and discovery of domestic and international experts and scholars, nowadays DSS has developed into the New Types of Decision Support Systems that data warehouse, on-line analytical processing and data mining combine.Its typical feature is the information obtaining aid decision from mass data.Data mining (DM, DataMining) is a special kind of skill extracting valuable knowledge from mass data.Constantly perfect along with data mining technology, data mining obtains in decision support field and applies more and more widely.These knowledge are that decision-making provides strong support.Radio and television individual character program commending, based on DSS, builds the model and method of dealing with problems, and by data mining technology digging user viewing behavior rule and the potential rating crowd of excavation.
The essence of individual character program commending sorts to the program that user watches, in this field, current existing method has the sort algorithm etc. under simple statistics algorithm, simple cascade clustering algorithm, Bayes network algorithm, multiple characteristics.The common issue that above several method exists is the sequence that only can realize watching user program, but can not provide different service for the user without feature, does not possess the ability of hiving off to viewer simultaneously.
Summary of the invention
The present invention is for solving the aforementioned problems in the prior a little and making, its object is to provide the individual character program commending method in a kind of radio data system and this radio data system, broadcast TV program can be recommended neatly according to the different demands of viewer, realize the function of individual character program commending.
For this reason, the invention provides a kind of radio data system, it comprises: input part, carries out various parameter needed for individual character program commending and various instruction for inputting above-mentioned radio data system; Programme information storage part, for storing information about various broadcast TV program and data; Analytic unit, is utilized the various parameter inputted by input part and the information about broadcast TV program read from above-mentioned programme information storage part, generates the individual character programme that will send, and determine the recommendation crowd as sending object; And recommendation information sending part, send above-mentioned individual character programme to the above-mentioned rating colony determined.
In addition, the present invention also provides the program commending method of the individual character in a kind of radio data system, this radio data system comprises input part, programme information storage part, analytic unit and recommendation information sending part, it is characterized in that, the method comprises the following steps: input above-mentioned radio data system by input part and carry out various parameter needed for individual character program commending and various instruction; Analytical procedure, is utilized the various parameter inputted by input part and the information about broadcast TV program read from above-mentioned programme information storage part, generates the individual character programme that will send, and determine the recommendation crowd as sending object; And by recommendation information sending part, send above-mentioned individual character programme to the above-mentioned rating colony determined.
Beneficial effect:
Present invention achieves the solution selecting individual character program commending method according to the different demand of radio and television user flexibly.The program category Threshold Analysis method provided or clustering method, can realize assisting program making business to stablize the loyal spectators of program, find the object of the potential spectators of program.User watched behavior analysis method, by the analysis of particular user viewing behavior, can realize the rating preference effectively holding user, recommends the object of individual character program.
Accompanying drawing explanation
Fig. 1 is the schematic diagram of the concrete structure representing the radio data system 100 that the present invention relates to.
Fig. 2 is the flow chart of the individual character program recommendation process represented performed by above-mentioned radio data system 100.
Fig. 3 is the flow chart of first example of the analytic process that the analytic unit 130 represented in above-mentioned radio data system 100 performs.
Fig. 4 is the flow chart of second example of the analytic process that the analytic unit 130 represented in above-mentioned radio data system 100 performs.
Embodiment
Below, the specific embodiment of the individual character program commending method in the radio data system and this system that the present invention relates to is described with reference to accompanying drawing.
Fig. 1 is the schematic diagram of the concrete structure representing the radio data system 100 that the present invention relates to.
As shown in Figure 1, radio data system 100 of the present invention comprises input part 110, programme information storage part 120, analytic unit 130, recommendation information sending part 140.
Wherein, input part 110 carries out data and the various instructions such as the various parameters needed for individual character program commending for inputting above-mentioned radio data system, and it can be keyboard, touch-screen, handwriting input device, mouse etc.
Programme information storage part 120 for storing the information about various broadcast TV program, the type of such as program, the time parameter of program, the various threshold values etc. preset.In addition, above-mentioned programme information storage part 120 can also store other data needed for above-mentioned radio data system 100 n-back test.These information and data can be stored in advance in above-mentioned programme information storage part 120, are stored in above-mentioned programme information storage part 120 after also can being inputted by input part 110.
Analytic unit 130 is for the information about broadcast TV program utilizing the data such as the various parameters that inputted by input part 110 and read from above-mentioned programme information storage part 120, analyzing and processing is carried out to above-mentioned various parameter and information, the individual character programme that generation will send, and determine the recommendation crowd as sending object.In the present invention, analytic unit 130 can utilize program category analytic approach (such as program category Threshold Analysis method, program category clustering methodology), viewing behavior analytic approach (such as rating individual behavior analytic approach, rating group behavior analytic approach) in any one, wherein, when utilizing program threshold type analytic approach or program category clustering methodology, recommendation crowd can be determined by calculating rating crowd occupation rate, when utilizing rating individual behavior analytic approach or rating group behavior analytic approach, the channel of recommended program list can be determined by calculating channel contributions rate, thus the individual character programme generated for recommending.Above-mentioned which analytic approach of analytic unit 130 choice for use, can be determined by the user instruction from input part 110.And the particular content about these analytic approachs will be discussed below in detail.
Recommendation information sending part 140, sends above-mentioned individual character programme by the mode such as short message mode or Email to the rating individuality determined or colony.At this, this recommendation information sending part 140 can be short message sending platform, also can be that Email sends platform.
Then, with reference to Fig. 2 illustrate above-mentioned radio data system institute 100 execution individual character program recommendation process.
First, user carries out data and the various instructions (step S211) such as the various parameters needed for individual character program commending by input part 110 input.At this, these parameters can comprise the parameter of user location, the type parameter of program and time parameter etc.Various instruction can comprise the instruction of the analytical method utilized for selection analysis unit 130, can also comprise the instruction of the send mode determining recommendation information sending part 140.The data such as above-mentioned parameter can be stored in programme information storage part 120, and analytic unit 130 also can be sent to use.
Then, in step S212, analytic unit 130 carries out analyzing and processing to above-mentioned various parameter and information, generates the individual character programme that will send, and determines the recommendation individuality as sending object or colony.At this, analytic unit 130 can determine to use which kind of analytic approach according to the instruction from input part 110.Analytic unit 130 determines the individual or colony of the rating as recommended, and generates the individual character programme that or colony individual to above-mentioned recommendation recommend.
Then, in step S213, above-mentioned recommendation information sending part 140 is according to selected send mode, and to fixed rating, individual or colony sends individual character programme.
It should be noted that, the term " crowd " recorded in this specification, can be independent rating individuality, the rating colony that also can be made up of multiple rating individuality.
Below, the analysis action performed by above-mentioned analytic unit 130 is described in detail with reference to Fig. 3 and Fig. 4.Fig. 3 is the flow chart of first example of the analytic process that the analytic unit 130 represented in above-mentioned radio data system 100 performs.Fig. 4 is the flow chart of second example of the analytic process that the analytic unit 130 represented in above-mentioned radio data system 100 performs.
(first case: program category analytic approach)
First, in step S311, carry out Selecting parameter.At this, above-mentioned parameter comprises regional parameters, program category parameter and time parameter.
In the present embodiment, regional parameters can select the data in any area existed in database, wherein, in units of above-mentioned data Shi Yishenghuo city DBMS.
In the present embodiment, program category parameter can select two-stage program classification or three grades of program classifications, and wherein the two-stage classification of program category comprises 4 classes, is respectively news controlling, amusement class program, educational program and service class program; Above-mentioned three grades of program classifications of above-mentioned two-stage program classification comprise 27 classes altogether, and wherein above-mentioned three grades of program classifications of above-mentioned news controlling are roundup news message program, classified news message program, Special Topics in Journalism class program, news talk show, world news class program, large-scale news program; Three grades of program classifications of above-mentioned amusement class program are TV play program, sports cast, film class program, variety show, music program, drama programs, game shows, reality TV show program, amusement talk. feature program, international amusement class program, large-scale entertainment; Above-mentioned three grades of program classifications of above-mentioned educational program are social education program, juvenile. young program, international education class program, large-scale then educational programs; Three grades of program classifications of above-mentioned service class program are Service Programmer, financing program, commercial paper program, national service class program, channel publicity. rating service packages, large-scale service packages.
In addition, in the present embodiment, in Duan Weiyi week analysis time selected by above-mentioned time parameter, the time period of recommended program list is next week of current time.
In step S312, generate individual character programme according to above-mentioned program category, and using the programme of this individual character programme as the above-mentioned program category in next week of current time.
In step S313, calculate rating duration according to above-mentioned regional parameters, program category parameter and time parameter.Above-mentioned rating duration can be obtained by following formula:
T = Σ i = 1 n T i
Wherein:
N represents the number of programs in selected program category;
T irepresent effective rating duration.
In the present embodiment, effective rating duration comprises following four kinds of situations:
Work as WR_Begin<TV_Begin, during WR_End<TV_End, effective rating duration is defined as WR_End-TV_Begin;
Work as TV_Begin<WR_Begin, during TV_End<WR_End, effective rating duration is defined as TV_End-WR_Begin;
Work as WR_Begin<TV_Begin, during TV_End<WR_End, effective rating duration is defined as TV_End-TV_Begin;
Work as TV_Begin<WR_Begin, during WR_End<TV_End, effective rating duration is defined as WR_End-WR_Begin.
Wherein, WR_Begin and WR_End represents the time (WR_Begin<WR_End) that the rating record satisfied condition starts and stops respectively;
TV_Begin and TV_End represents that certain program broadcasts the time (TV_Begin<TV_End) starting and stop respectively;
Step S314, the above-mentioned rating duration according to calculating carries out rating population analysis.
In the present embodiment, analyze any one that above-mentioned rating crowd can use in program category Threshold Analysis method or program category clustering methodology.
Above-mentioned program category Threshold Analysis method comprises the following steps:
Utilize the maximum rating duration of two threshold value i and j(0<i<j< preset), rating crowd is carried out following classification:
As T≤i, be loss family, namely seldom watch the potential family of above-mentioned program category;
As i<T<j, be average family, namely paying close attention to more to the above-mentioned type program, is not again the average family extremely made earnest efforts;
As j≤T, be loyal family, namely to the loyal family that the above-mentioned type program is extremely paid close attention to; Wherein, T is above-mentioned rating duration.
In addition, above-mentioned program category clustering methodology comprises the following steps: setting above-mentioned rating duration number to be clustered is n(n>0), crowd's cluster number k(0<k≤n of above-mentioned rating duration),
Steps A: above-mentioned average as the initial mean value of each bunch, and is defined as the assembly average of each bunch of above-mentioned rating duration by random selecting k above-mentioned rating duration,
Step B: definition mean square error E
E = &Sigma; i = 1 k &Sigma; T &Element; Ci | T - m i | 2
Wherein:
T represents above-mentioned rating duration;
C irepresent certain bunch;
M irepresent bunch C ithe average of above-mentioned rating duration;
According to as defined above, calculate the above-mentioned mean square error E of each the above-mentioned rating duration in each bunch, and each above-mentioned rating duration is assigned to the most similar bunch, namely with the distance of bunch average minimum bunch;
Step C: for upgrade after bunch, calculate the average of above-mentioned rating duration in each bunch;
Step D: repeat step B to step C, until bunch no longer to change after upgrading, then obtains the k after above-mentioned cluster analysis process bunch, i.e. k class crowd;
Sort according to each bunch of average is descending, define successively each bunch be rank 1 family, rank 2 family ... wherein rank 1 family is the highest to this type of program informativeness, and after this rank informativeness successively decreases successively.
Then, in step S315, rating crowd occupation rate is calculated according to the above-mentioned rating crowd that above-mentioned Threshold Analysis draws.Above-mentioned rating crowd occupation rate can utilize following formula to obtain:
PER = N &prime; N &times; 100 %
Wherein:
N' represents the rating amount (0≤N') of certain class crowd;
N represents the total amount of rating (N'≤N) in selected area.
In step S316, with reference to above-mentioned rating crowd occupation rate, determine the recommendation crowd as the recommended sending individual character programme.Wherein, above-mentioned recommendation crowd can be the combination of any crowd or any several crowd.Like this, just generated the individual character programme that will send by analytic unit 130, and determine the recommendation crowd as sending object.
(second case: viewing behavior analytic approach)
Viewing behavior analytic approach comprises individual viewing behavior analytic approach and colony's viewing behavior analytic approach, and wherein, individual viewing behavior analytic approach selects the audience information of a certain individuality to carry out analytical calculation, obtains the recommendation channel to this individual recommended program list; Colony's viewing behavior analytic approach selects the audience information of a certain rating colony to carry out analytical calculation, obtains the recommendation channel to this colony's recommended program list.Below, the concrete analysis step of above-mentioned individual viewing behavior analytic approach and colony's viewing behavior analytic approach is described respectively with reference to Fig. 4.
1, individual viewing behavior analytic approach
Below, the detailed process of individual viewing behavior analytic approach is described with reference to Fig. 4.
First, in step S411, select the particular user as analytic target and time parameter, and extract the information of above-mentioned particular user.
Then, in step S412, calculate the rating duration of above-mentioned particular user, above-mentioned rating duration can be obtained by following formula:
T sl = &Sigma; l = 1 n 4 &Sigma; k = 1 n 3 &Sigma; j = 1 n 2 &Sigma; i = 1 n 1 T i , j , k , l
T s 2 = &Sigma; k = 1 n 3 &Sigma; j = 1 n 2 &Sigma; i = 1 n 1 T i , j , k
Wherein:
T s1represent that above-mentioned particular user watches the total duration of the rating of all programs;
T s2represent that above-mentioned particular user watches the total duration of the rating of certain class program;
N 1expression one day, above-mentioned particular user watched the number of programs of certain channel class program;
N 2expression one day, above-mentioned particular user watched the number of channels of certain class program;
N 3represent the input number of days analyzing the above-mentioned 3rd rating duration of particular user;
N 4represent that above-mentioned particular user watches program category sum;
T i, j, k, land T i, j, krepresent the effective rating duration watching certain concrete program.
Then, in step S413, utilize following formula to calculate the rating preference of this particular user:
And by above-mentioned rating preference by descending sequence, to the program category that above-mentioned particular user recommends rating preference maximum.
In step S414, according to the result of calculation of above-mentioned rating preference, choose the program category that above-mentioned rating preference is maximum, utilize the result calculating above-mentioned rating duration and obtain, utilize following formula to calculate channel contributions rate:
&psi; s = T s 3 T s 2
Wherein:
T s 3 = &Sigma; k = 1 n 3 &Sigma; i = 1 n 1 T i , k
T s3represent that the rating duration of certain such program of channel is watched at the concrete family of above-mentioned use;
N 1expression one day, above-mentioned particular user watched the number of programs of certain such program of channel;
N 3represent the input number of days of the rating duration analyzing above-mentioned particular user;
T i,krepresent the effective rating duration watching certain concrete program.
Then, in step S415, above-mentioned channel contributions rate is sorted, according to above-mentioned channel contributions rate and the programme of above-mentioned rating preference generating recommendations, determine the channel recommending above-mentioned recommended program list, wherein, above-mentioned channel can select single channel or the combinations of channels of arbitrary channel contribution rate.
2, colony's viewing behavior analytic approach
Colony's viewing behavior analytic approach is similar with the general steps of individual viewing behavior analytic approach, therefore, below the same detailed process that colony's viewing behavior analytic approach is described with reference to Fig. 4.
First, in step S411, select the rating group types as analytic target and time parameter, and extract the information of this rating colony, such as, group classification comprise real estate/building, service trade, industry/geology, radio and television/culture and arts, computer (IT/ the Internet), traffic/transport, education/training, finance (bank/security/insurance), hotel/tourism/food and drink, trade/import and export, media/advertisement/consulting, agricultural/aquatic products, retirement, student, medical treatment/health care/pharmacy, government bodies and other etc. totally 17 class crowds.
Then, in step S412, the rating duration of the above-mentioned rating colony selected by following formula calculating is utilized:
T cl = &Sigma; p = 1 n 5 &Sigma; l = 1 n 4 &Sigma; k = 1 n 3 &Sigma; j = 1 n 2 &Sigma; i = 1 n 1 T i , j , k , l , p
T c 2 = &Sigma; p = 1 n 5 &Sigma; k = 1 n 3 &Sigma; j = 1 n 2 &Sigma; i = 1 n 1 T i , j , k , p
Wherein:
T c1represent that the total duration of the rating of all programs is watched by above-mentioned rating colony;
T c2represent that the total duration of the rating of certain class program is watched by above-mentioned rating colony;
N 1expression one day, the number of programs of certain class program of certain channel was watched by above-mentioned rating colony;
N 2expression one day, the number of channels of certain class program was watched by above-mentioned rating colony;
N 3represent the input number of days analyzing above-mentioned rating colony rating duration;
N 4represent that the sum of program category is watched by above-mentioned rating colony;
N 5represent the individual number of rating that above-mentioned rating colony comprises;
T i, j, k, l, pand T i, j, k, prepresent the effective rating duration watching certain concrete program.
Then, in step S413, following formula is utilized to calculate the rating preference of above-mentioned rating colony
Further, by rating preference by the sequence of descending order, to the program category that group of subscribers recommends rating preference maximum.
Then, in step S414, according to the above-mentioned rating preference calculated, choose the program category that rating preference is maximum, and utilize the result that rating duration calculation obtains, calculate channel contributions rate Ψ based on following formula c:
&psi; c = T c 3 T c 2
Wherein:
T c 3 = &Sigma; p = 1 n 5 &Sigma; k = 1 n 3 &Sigma; i = 1 n 1 T i , k , p
T c3represent that the rating duration of certain such program of channel is watched by above-mentioned rating colony;
N 1expression one day, the number of programs of such program was watched by above-mentioned rating colony;
N 3represent the input number of days analyzing above-mentioned rating colony rating duration;
N 5represent the individual number comprised in above-mentioned rating colony;
T i, k, prepresent the effective rating duration watching certain concrete program.
Finally, in step S415, channel contributions rate sorted, income value larger channel contributions rate is higher, and, according to channel contributions rate, select the channel of recommended program list, wherein, single channel or the combinations of channels of arbitrary channel contribution rate can be selected.
According to said method, individual character program commending method can be selected flexibly, to reach the object of promotion individual character program according to the different demand of radio and television user.
Preferably, the above-mentioned user of the present embodiment can be individual consumer, also can be group of subscribers.
According to as above above-mentioned the present invention, achieve the solution selecting individual character program commending method according to the different demand of radio and television user flexibly.The program category Threshold Analysis method provided or clustering method, can realize assisting program making business to stablize the loyal spectators of program, find the object of the potential spectators of program.User watched behavior analysis method, by the analysis of particular user viewing behavior, can realize the rating preference effectively holding user, recommends the object of individual character program.
Under above-mentioned instruction of the present invention, those skilled in the art can improve the individual character program commending method of radio and television provided by the present invention and system on the basis of above-described embodiment, and these improvement all drop in protection scope of the present invention.It will be understood by those skilled in the art that above-mentioned specific descriptions just explain object of the present invention better, protection scope of the present invention is by claim and equivalents thereof.

Claims (6)

1. a radio data system, is characterized in that, comprising:
Input part, carries out various parameter needed for individual character program commending and various instruction for inputting above-mentioned radio data system;
Programme information storage part, for storing information about various broadcast TV program and data;
Analytic unit, is utilized the various parameter inputted by input part and the information about broadcast TV program read from above-mentioned programme information storage part, generates the individual character programme that will send, and determine the recommendation crowd as sending object; And
Recommendation information sending part, to the above-mentioned individual character programme of above-mentioned referrer's pocket transmission determined, wherein,
Described analytic unit performs following analytic process:
(1) regional parameters, program category parameter and time parameter is selected;
(2) individual character programme is generated according to above-mentioned program category parameter, and the programme of above-mentioned program category as next week of current time;
(3) calculate rating duration according to above-mentioned regional parameters, program category parameter and time parameter, above-mentioned rating duration is obtained by following formula:
T = &Sigma; i = 1 n T i
Wherein:
N represents the number of programs in selected program category;
T irepresent effective rating duration;
(4) according to the above-mentioned rating duration calculated, rating crowd is calculated;
(5) calculate rating crowd occupation rate according to above-mentioned rating crowd, rating crowd occupation rate is obtained by following formula:
P E R = N &prime; N &times; 100 %
Wherein:
N' represents the rating amount of certain class crowd, 0≤N';
N represents the total amount of rating in selected area, N'≤N;
(6) with reference to above-mentioned rating crowd occupation rate, recommendation crowd is determined.
2. radio data system as claimed in claim 1, it is characterized in that, the step of above-mentioned calculating above-mentioned rating crowd is further comprising the steps of:
Preset two threshold value i and j, wherein the maximum rating duration of 0<i<j<, according to above-mentioned threshold value, rating crowd classified as follows:
As T≤i, being loss family, as i<T<j, is average family, as j≤T, is loyal family, and wherein, T is above-mentioned rating duration.
3. radio data system as claimed in claim 1, it is characterized in that, the step calculating above-mentioned rating crowd is further comprising the steps of:
Setting above-mentioned rating duration number to be clustered is n, crowd's cluster number k of above-mentioned rating duration, wherein n>0,0<k≤n;
Steps A: above-mentioned average as the initial mean value of each bunch, and is defined as the assembly average of each bunch of above-mentioned rating duration by random selecting k above-mentioned rating duration,
Step B: definition mean square error E
E = &Sigma; i = 1 k &Sigma; T &Element; C i | T - m i | 2
Wherein, T represents above-mentioned rating duration, C irepresent certain bunch, m irepresent bunch C ithe average of above-mentioned rating duration;
According to as defined above, calculate the above-mentioned mean square error E of each the above-mentioned rating duration in each bunch, and each above-mentioned rating duration is assigned to the most similar bunch, namely with the distance of bunch average minimum bunch;
Step C: for upgrade after bunch, calculate the average of above-mentioned rating duration in each bunch;
Step D: repeat above-mentioned steps B to step C, until bunch no longer to change after upgrading, then obtains the k after above-mentioned cluster analysis process bunch, i.e. k class crowd;
Sort according to each bunch of average is descending, define successively each bunch be rank 1 family, rank 2 family ... wherein rank 1 family is the highest to this type of program informativeness, and after this rank informativeness successively decreases successively.
4. the individual character program commending method in radio data system, this radio data system comprises input part, programme information storage part, analytic unit and recommendation information sending part, and it is characterized in that, the method comprises the following steps:
Input above-mentioned radio data system by input part and carry out various parameter needed for individual character program commending and various instruction;
Analytical procedure, is utilized the various parameter inputted by input part and the information about broadcast TV program read from above-mentioned programme information storage part, generates the individual character programme that will send, and determine the recommendation crowd as sending object; And by recommendation information sending part, to the above-mentioned individual character programme of above-mentioned referrer's pocket transmission determined, wherein,
Described analytical procedure performs following analytic process:
(1) regional parameters, program category parameter and time parameter is selected;
(2) individual character programme is generated according to above-mentioned program category parameter, and the programme of above-mentioned program category as next week of current time;
(3) calculate rating duration according to above-mentioned regional parameters, program category parameter and time parameter, above-mentioned rating duration is obtained by following formula:
T = &Sigma; i = 1 n T i
Wherein:
N represents the number of programs in selected program category;
T irepresent effective rating duration;
(4) according to the above-mentioned rating duration calculated, rating crowd is calculated;
(5) calculate rating crowd occupation rate according to above-mentioned rating crowd, rating crowd occupation rate is obtained by following formula:
P E R = N &prime; N &times; 100 %
Wherein:
N' represents the rating amount of certain class crowd, 0≤N';
N represents the total amount of rating in selected area, N'≤N;
(6) with reference to above-mentioned rating crowd occupation rate, recommendation crowd is determined.
5. individual character program commending method as claimed in claim 4, it is characterized in that, the step of above-mentioned calculating above-mentioned rating crowd is further comprising the steps of:
Preset two threshold value i and j, wherein the maximum rating duration of 0<i<j<, according to above-mentioned threshold value, rating crowd classified as follows:
As T≤i, being loss family, as i<T<j, is average family, as j≤T, is loyal family, and wherein, T is above-mentioned rating duration.
6. individual character program commending method as claimed in claim 4, it is characterized in that, the step calculating above-mentioned rating crowd is further comprising the steps of:
Setting above-mentioned rating duration number to be clustered is n, crowd's cluster number k of above-mentioned rating duration, wherein n>0,0<k≤n;
Steps A: above-mentioned average as the initial mean value of each bunch, and is defined as the assembly average of each bunch of above-mentioned rating duration by random selecting k above-mentioned rating duration,
Step B: definition mean square error E
E = &Sigma; i = 1 k &Sigma; T &Element; C i | T - m i | 2
Wherein, T represents above-mentioned rating duration, C irepresent certain bunch, m irepresent bunch C ithe average of above-mentioned rating duration;
According to as defined above, calculate the above-mentioned mean square error E of each the above-mentioned rating duration in each bunch, and each above-mentioned rating duration is assigned to the most similar bunch, namely with the distance of bunch average minimum bunch;
Step C: for upgrade after bunch, calculate the average of above-mentioned rating duration in each bunch;
Step D: repeat above-mentioned steps B to step C, until bunch no longer to change after upgrading, then obtains the k after above-mentioned cluster analysis process bunch, i.e. k class crowd;
Sort according to each bunch of average is descending, define successively each bunch be rank 1 family, rank 2 family ... wherein rank 1 family is the highest to this type of program informativeness, and after this rank informativeness successively decreases successively.
CN201310272576.9A 2012-09-17 2013-07-01 Individual character program commending method in radio data system and this system Active CN103338403B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201310272576.9A CN103338403B (en) 2012-09-17 2013-07-01 Individual character program commending method in radio data system and this system

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
CN2012103425409 2012-09-17
CN201210342540.9 2012-09-17
CN201210342540 2012-09-17
CN201310272576.9A CN103338403B (en) 2012-09-17 2013-07-01 Individual character program commending method in radio data system and this system

Related Child Applications (2)

Application Number Title Priority Date Filing Date
CN201610112192.4A Division CN105681908A (en) 2012-09-17 2013-07-01 Broadcast television system based on individual watching behaviour and personalized programme recommendation method thereof
CN201610112159.1A Division CN105744370A (en) 2012-09-17 2013-07-01 Radio and television system based on group viewing behaviors and personalized program recommendation method thereof

Publications (2)

Publication Number Publication Date
CN103338403A CN103338403A (en) 2013-10-02
CN103338403B true CN103338403B (en) 2016-03-16

Family

ID=49246487

Family Applications (3)

Application Number Title Priority Date Filing Date
CN201610112159.1A Pending CN105744370A (en) 2012-09-17 2013-07-01 Radio and television system based on group viewing behaviors and personalized program recommendation method thereof
CN201310272576.9A Active CN103338403B (en) 2012-09-17 2013-07-01 Individual character program commending method in radio data system and this system
CN201610112192.4A Pending CN105681908A (en) 2012-09-17 2013-07-01 Broadcast television system based on individual watching behaviour and personalized programme recommendation method thereof

Family Applications Before (1)

Application Number Title Priority Date Filing Date
CN201610112159.1A Pending CN105744370A (en) 2012-09-17 2013-07-01 Radio and television system based on group viewing behaviors and personalized program recommendation method thereof

Family Applications After (1)

Application Number Title Priority Date Filing Date
CN201610112192.4A Pending CN105681908A (en) 2012-09-17 2013-07-01 Broadcast television system based on individual watching behaviour and personalized programme recommendation method thereof

Country Status (1)

Country Link
CN (3) CN105744370A (en)

Families Citing this family (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104410874A (en) * 2014-11-19 2015-03-11 北京国双科技有限公司 A method, a device, and a system for detecting video viscosity information
CN104333777B (en) * 2014-11-28 2017-06-20 中国传媒大学 A kind of broadcast TV program list formulates system and method
CN104602040B (en) * 2014-11-28 2017-08-29 中国传媒大学 System and method is formulated in a kind of programme
CN104918117B (en) * 2015-03-24 2018-10-19 四川长虹电器股份有限公司 A kind of smart television advertisement and user tag recommend method
CN104768070A (en) * 2015-04-23 2015-07-08 天脉聚源(北京)传媒科技有限公司 Region-based channel sorting method and device
CN106028156B (en) * 2016-06-24 2017-03-01 合肥工业大学 Televiewer's interest modeling method and system
CN106604137B (en) * 2016-12-29 2020-06-12 Tcl科技集团股份有限公司 Method and device for predicting video watching duration
CN107241623B (en) * 2017-05-26 2019-08-02 中国传媒大学 The user watched behavior prediction method and system of radio and television
WO2019014834A1 (en) * 2017-07-18 2019-01-24 深圳市智晟达科技有限公司 Method for counting watching time, and digital television
CN108470050A (en) * 2018-03-09 2018-08-31 吉林农业大学 A kind of space-time of facing agricultural internet web resource recommends method and system
CN108366276B (en) * 2018-03-16 2020-05-01 中国传媒大学 Viewing preference analysis method and system
CN108259982B (en) * 2018-04-13 2020-09-08 中广热点云科技有限公司 Television channel control method and system arranged on mobile terminal
CN108737856B (en) * 2018-04-26 2020-03-20 西北大学 Social relation perception IPTV user behavior modeling and program recommendation method
CN108769817A (en) * 2018-05-31 2018-11-06 深圳市路通网络技术有限公司 Program commending method and system
CN111222923B (en) * 2020-01-13 2023-12-15 秒针信息技术有限公司 Method and device for judging potential clients, electronic equipment and storage medium

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1608173A1 (en) * 2004-06-14 2005-12-21 Sony Corporation Program information processing system, program information management server, program information operation terminal, and computer program
CN101489107A (en) * 2009-01-21 2009-07-22 华东师范大学 Collaborative filtering recommendation method based on population attribute keyword vector
CN102056018A (en) * 2010-11-26 2011-05-11 Tcl集团股份有限公司 Method and system for providing TV guide and method for providing program-requesting information

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1608173A1 (en) * 2004-06-14 2005-12-21 Sony Corporation Program information processing system, program information management server, program information operation terminal, and computer program
CN101489107A (en) * 2009-01-21 2009-07-22 华东师范大学 Collaborative filtering recommendation method based on population attribute keyword vector
CN102056018A (en) * 2010-11-26 2011-05-11 Tcl集团股份有限公司 Method and system for providing TV guide and method for providing program-requesting information

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
《基于用户聚类的播客节目推荐》;陈超 等;《计算机应用与软件》;20090331;第26卷(第3期);第1-3页的第1-3部分 *

Also Published As

Publication number Publication date
CN105744370A (en) 2016-07-06
CN105681908A (en) 2016-06-15
CN103338403A (en) 2013-10-02

Similar Documents

Publication Publication Date Title
CN103338403B (en) Individual character program commending method in radio data system and this system
Ksiazek A network analytic approach to understanding cross-platform audience behavior
Weaver et al. The (non) violent world of YouTube: Content trends in web video
CN106062730A (en) Systems and methods for actively composing content for use in continuous social communication
CN105516810A (en) Television user family member analysis method based on LDA (Latent Dirichlet Allocation) model
Lane et al. Fostering smarter colleges and universities
CN110474944B (en) Network information processing method, device and storage medium
CN106851349A (en) Based on magnanimity across the live recommendation method for shielding viewing behavior data
CN113742567B (en) Recommendation method and device for multimedia resources, electronic equipment and storage medium
CN104572982A (en) Personalized recommendation method and system based on question guide
CN109597899A (en) The optimization method of media personalized recommendation system
Aryal et al. MoocRec: Learning styles-oriented MOOC recommender and search engine
Kenix The future of alternative media?
Bogaert et al. Predicting Self‐declared movie watching behavior using Facebook data and Information‐Fusion sensitivity analysis
Hidayat et al. Data-driven journalism based on big data analytics: A model development from Indonesia’s experience
Okwu Characterizing farmer users and nonusers of mass media as channels of agricultural information in Benue State, Nigeria
KR102559449B1 (en) Server and method for providing charge-free lerning curation service
Faris et al. Preparing a Big Data Utilization Communication Strategy
Xie BDA of the Dissemination of Opera in the Internet Self-Media Environment
Choudhury et al. Impact of artificial intelligence (AI) in the media and entertainment industry
Dutta et al. Internet skills as an influencer for adoption of digital innovations in a technologically emerging nation: India
Ptaszek From algorithmic surveillance to algorithmic awareness: Media education in the context of new media economics and invisible technologies
Badiru Nollywood’s Coverage and Framing of Agriculture in the Transformation Agenda Period (2013–2014) in Nigeria
Yuan et al. A Systematic Review of New Media Influencing People’s Attitude and Cognition to GMF
Adetunji et al. Design of a Movie Review Rating Prediction (MR2P) Algorithm

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant